United States – Subsidies on Upland Cotton

(WT/DS267)

Executive Summary of the Rebuttal Submission of the United States of America

September 1, 2003 Introduction and Overview

1. The comparison under the Peace Clause proviso in Article 13(b)(ii) must be made with respect to the support as “decided” by those measures. In the case of the challenged U.S. measures, the support was decided in terms of a rate, not an amount of budgetary outlay. The rate of support decided during marketing year 1992 was 72.9 cents per pound of upland cotton; the rate of support granted for the 1999-2001 crops was only 51.92 cents per pound; and the rate of support that measures grant for the 2002 crop is only 52 cents per pound. Thus, in no marketing year from 1999 through 2002 have U.S. measures breached the Peace Clause.1

2. Brazil has claimed that additional “decisions” by the United States during the 1992 marketing year to impose a 10 percent acreage reduction program and 15 percent “normal flex acres” reduced the level of support below 72.9 cents per pound. However, the 72.9 cents per pound rate of support most accurately expresses the revenue ensured by the United States to upland cotton producers. Even on the unrealistic assumption that these program elements reduced the level of support by 10 and 15 percent, respectively (that is, the maximum theoretical effect these program elements could have had), the 1992 rate of support would still be 67.625 cents per pound, well above the levels for marketing years 1999-2001 and 2002.

3. Although such a comparison would not conform to the text, the result of the Peace Clause comparison is no different if one compares the support via an Aggregate Measurement of Support calculation. Using the price gap methodology (as provided under Annex 3 of the Agriculture Agreement) for U.S. price-based deficiency payments and marketing loan payments, the upland cotton Aggregate Measurement of Support (in U.S. $, millions) for these years is MY1992: 1,079; MY1999: 717; MY2000: 484; MY2001: 264; MY2002: 205. Again, in no marketing year from 1999 through 2002 have U.S. measures breached the Peace Clause.

4. Finally, the analysis presented by Brazil’s expert at the first panel meeting actually supports the United States, not Brazil. Removing the non-product-specific support that Brazil erroneously tries to pass off as support to upland cotton, Brazil’s own expert calculates the total support per unit (cents/lb.) as MY1992: 60.05; MY1999: 53.79; MY2000: 55.09; MY2001: 52.82; MY2002: 56.32. Again, in no marketing year from 1999 through 2002 have U.S. measures breached the Peace Clause.

1 Brazil has asserted that the United States’ approach does not provide any way of taking Step 2 payments into account. Because the availability of Step 2 payments is contingent on certain price conditions existing during the marketing year, the level of support decided must relate to the payment parameters. These have remained the same for Step 2, with the exception of the suspension, through 2006, of the 1.25 cent price difference threshold and payment availability at slightly higher market prices. However, because Step 2 merely provides an alternative avenue of providing support (through processors rather than directly to producers), these minor adjustments do not alter the revenue ensured for producers by the marketing loan rate of 52 cents per pound. In addition, these minor adjustments cannot overcome the greater than 20 cents per pound difference in product-specific support between marketing years 1992 and 1999-2002. Similarly, and without prejudice to whether these measures are within the Panel’s terms of reference, we note that cottonseed payments in 1999, 2000, and 2002 ranged in value between 0.6 to 2.3 cents per pound (factoring expenditures over production); thus, they too do not materially affect the comparison between marketing year 1992 and any other year. United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 2

5. Thus, whether gauged via the rate of support decided by U.S. measures (whether or not adjusted for the acreage reduction program and normal flex acres), or via the AMS for upland cotton (calculated through a price gap methodology), or via the calculations of Brazil’s expert (limited to product-specific support), the result is exactly the same: in no marketing year from 1999 through 2002 have U.S. measures breached the Peace Clause.

U.S. Green Box Measures Are “Exempt from Actions” Pursuant to Article 13(a)(ii)

6. A measure shall be deemed to meet the “fundamental requirement” of the first sentence of Annex 2 if it meets the basic criteria of the second sentence plus any applicable policy-specific criteria. As suggested by the use of the word “fundamental” (“from which others are derived”) and the structure of Annex 2 (that is, beginning the second sentence with the word “accordingly”), compliance with the requirement (“something called for or demanded”) of the first sentence will be demonstrated by conforming to the basic criteria of the second sentence plus the applicable policy-specific criteria of paragraphs 6 through 13.

7. Direct Payments: Eligibility for direct payments under the 2002 Act is based on criteria in a “defined and fixed base period” (paragraph 6(a)) in the ordinary meaning of those terms: a base period that is “definite” (set out in the 2002 Act) and “stationary or unchanging in a relative position” (does not change in relative position for the six-year duration of the 2002 Act).

8. Paragraph 6(a) establishes that eligibility for payments under a decoupled income support measure shall be determined by clearly-defined criteria in “a defined and fixed base period,” not “the base period” (as in paragraph 9 of Annex 3, which is defined in that same paragraph as “the years 1986 to 1988”). Brazil’s reading of “a defined and fixed base period” would read into that text the term “unchanging,” language Brazil has proposed in the ongoing WTO negotiations but is not currently found in the Agreement.

9. Annex 2, by its terms, sets out the fundamental requirement and basic and (if applicable) policy-specific criteria to which green box “domestic support measures” must conform. Other provisions in the Agreement similarly establish that the criteria set out in Annex 2 apply to “domestic support measures.” Thus, with respect to a given decoupled income support measure, eligibility for payments must be determined by criteria in a “defined and fixed base period.”

10. Brazil argues that a new decoupled income support measure must be based on the same base period as a previous measure if the new measure “is essentially the same” or “[i]f the structure, design, and eligibility criteria have not significantly changed.” There is no provision in Annex 2 or the that supports Brazil’s approach. It is thus irrelevant whether two decoupled income support measures are “essentially the same.” United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 3

11. Brazil would read paragraph 6(b) as requiring a Member to make support available for any type of production; a Member could not preclude a recipient from producing certain crops.2 While direct payments are reduced if certain crops are produced, a recipient need not produce any “type of” crop in particular in order to receive the full payment for which a farm is eligible; the recipient need merely refrain from producing the forbidden fruit or vegetable. Thus, it is not any “type . . . of production . . . undertaken by the producer” that results in the full direct payment but rather production not undertaken by the producer – that is, ceasing certain production.

12. Production Flexibility Contract Payments: Production flexibility contract payments (now expired) were made with respect to farm acreage that was devoted to agricultural production in the past, including acreage previously devoted to upland cotton production. The payments, however, were made regardless of whether upland cotton was produced on those acres or whether anything was produced at all. As with direct payments, because production flexibility contract payments were decoupled from production, they met the five policy-specific criteria set out in paragraph 6 for decoupled income support measures.

13. Brazil Has Not Shown U.S. Green Box Measures Are More than Minimally Trade- or Production-Distorting: Brazil has failed to make a prima facie case that U.S. green box measures do not satisfy the fundamental requirement of Annex 2.3 In fact, Brazil’s “evidence” consists simply of selectively quoting and emphasizing conceptual and theoretical statements from the economic literature. None of the papers Brazil cites concludes that these payments in particular, or decoupled income support measures in general, have more than “minimal[] trade-distorting effects or effects on production.”

14. The Agreement on Agriculture does not define a numerical threshold on what degree of effects will be considered “minimal[] trade-distorting effects or effects on production.” However, given that no study has found that these payments have effects on production of more than one percent, it would appear that direct payments have and production flexibility contract payments had no more than “minimal[] trade-distorting effects or effects on production.” Thus, not only has Brazil failed to present a prima facie case, but the United States has affirmatively shown that these payments satisfy the fundamental requirement of the first sentence of Annex 2.

2 Brazil’s reading would also seemingly require a Member to make payments even if the recipient’s production was illegal – for example, the production of narcotic crops such as opium poppy or the production of unapproved biotech varieties or environmentally damaging production (for example, planting on converted rain forest or wetlands) – because, under Brazil’s approach, by reducing or eliminating payments for any of these production activities, a decoupled income support measure could be understood to base or relate the amount of payment to the “type” of production undertaken. 3 If, as Brazil has argued, the first sentence is “fundamental” and has independent force, then presumably if a measure meets that “fundamental requirement,” it will be deemed to be green box, irrespective of whether it meets the subordinate basic and policy-specific criteria. Thus, on Brazil’s reading, if a measure does not conform to the criteria in Annex 2, it still could meet the “fundamental requirement,” and the complaining party would bear the burden of proof to demonstrate a measure’s inconsistency with that provision. United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 4

U.S. Non-Green Box Domestic Support Measures Are Not in Breach of Article 13(b)(ii)

15. Peace Clause Proviso – Support Was “Decided” During Marketing Year 1992 Using a Rate, Not a Budgetary Outlay: The Peace Clause proviso requires a comparison to the product-specific support “decided” during the 1992 marketing year. A Member cannot “decide” world market prices or actual production or any other element outside a government’s control. Yet Brazil would read the Peace Clause as though Members were omnipotent and could “decide” every factor influencing support.

16. Brazil lists nine different “decisions taken by the United States in relation to MY 1992 upland cotton support programs.” At least three of these “decisions” relate to the rate of support and not a single decision relating to budgetary outlays or market prices. Thus, Brazil’s own answer confirms that the proper analysis of the support “decided” by U.S. measures is to look to the terms of the U.S. measures, which set a rate of support.

17. The use of the term “grant” in the Peace Clause proviso with respect to challenged measures does not compel an examination of budgetary outlays. The ordinary meaning of “grant” is to “bestow as a favour” or “give or confer (a possession, a right, etc.) formally.” Thus, the use of the term “grant” would permit an evaluation of the rate of support that challenged measures “give or confer . . . formally.” Members did not choose to use the word “granted” in place of “decided,” and a valid interpretation must make sense of that choice rather than reading it out of the Agreement. In addition, had Members intended the Peace Clause comparison to be made solely on the basis of budgetary outlays, they could have used that term, which is a defined term in Article 1(c) and used frequently in the Agreement.

18. Peace Clause Proviso – “Support to a Specific Commodity” Means Product-Specific Support: The phrase “support to a specific commodity” means “product-specific support.” That the Peace Clause does not use the phrase “product-specific support” is neither surprising nor telling. The basic definition of product-specific support is given in Article 1(a), as “support . . . provided for an agricultural product in favour of the producers of the basic agricultural product.” Article 1(h) also refers to the concept but does not use the exact phrase “product-specific support”; in fact, the language this provision uses ("support for basic agricultural products") is strikingly similar to the Peace Clause proviso ("support to a specific commodity"). Neither Article 1(a) nor 1(h) even uses the term “specific” whereas the Peace Clause contains all three elements of that phrase (product, specific, and support).

19. Brazil Simply Ignores the Definition of Product-Specific Support in the Agreement on Agriculture: Brazil argues that certain challenged U.S. measures are not “non-product- specific” and therefore must be “support to a specific commodity.” Brazil focuses on the definition of “non-product-specific” support in Article 1(a) but simply fails to interpret that definition in light of the definition of product-specific support that immediately precedes it. The universe of domestic support measures under Article 1(a) consists of product-specific support and non-product-specific support; these two parts must be read together and in harmony. United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 5

20. The definition of product-specific support consists of two elements: First, the support must be provided “for an agricultural product,” that is, the subsidy is given “in favour of” a product and not in respect of criteria not related to the product or in respect of multiple products. Second, such support is “in favour of the producers of the basic agricultural product,” which suggests that subsidy benefits those who produce the product – that is, production is necessary for the support to be received. Both of these elements must be present for support to be product- specific since, should either be missing, the definition would not be satisfied.

21. The second category of support in Article 1(a) is defined as “non-product-specific support provided in favour of agricultural producers in general.” The ordinary meaning of “in general” is “in general terms, generally.” Non-product-specific support cannot be interpreted as support provided “for an agricultural product in favor of the producers of the basic agricultural product” because to do so would reduce the first half of the Article 1(a) definition to redundancy or inutility. Thus, non-product-specific support is support in favor of agricultural producers “generally” – that is, a residual category of support covering those measures that do not fall within the more detailed criteria set out in the definition of product-specific support.

22. Counter-Cyclical Payments Are Non-Product-Specific Support: Counter-cyclical payments are non-product-specific support. The payment formula for counter-cyclical payments demonstrates that these payments are not “provided for an agricultural product” because a recipient need not currently produce upland cotton (or any other crop) to receive payment. In addition, it is not “the producers of the basic agricultural product” – that is, current upland cotton growers – that are entitled to receive the counter-cyclical payments but rather persons (farmers and landowners) on farm acres with past histories of producing covered commodities, including upland cotton, during the base period. Thus, counter-cyclical payments satisfy neither element of the definition of product-specific support and do not form part of the Peace Clause comparison.

23. Despite Brazil’s attempts to mischaracterize the two as similar, counter-cyclical payments and deficiency payments differ in crucial respects. To receive a deficiency payment, a producer was required to plant upland cotton for harvest and would be paid on the acres planted to upland cotton for harvest up to the maximum payment acreage. Thus, deficiency payments were support for an agricultural product (upland cotton) in favor of the producers of the product. By contrast, to receive the counter-cyclical payment a person with “upland cotton base acres” need not plant for harvest or produce upland cotton (nor any other crop nor any crop at all). Thus, counter- cyclical payments do not provide support for “an agricultural product” in favor of “the producers” of the basic agricultural product and do not form part of the Peace Clause comparison under the proviso in Article 13(b)(ii).

24. Crop Insurance Payments Provide Non-Product-Specific Support: Crop insurance is not support “provided for an agricultural product.” For marketing year 2002, crop insurance payments are available to approximately 100 agricultural commodities, representing approximately 80 percent of U.S. area planted and greater than 85 percent of the value of all U.S. United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 6 crops. Support which is provided to a number of crops is not “support to a specific commodity”; it is ‘support to several commodities’ or ‘support to more than one commodity’ and does not form part of the Peace Clause comparison. The United States notifies crop insurance as non- product-specific “amber box” domestic support subject to U.S. reduction commitments. No WTO Member has notified crop insurance programs as product-specific; in fact, , Canada, the EC, and Japan have notified crop insurance programs as non-product-specific support. The United States is not aware of any other Member’s crop insurance program that has as broad product coverage as the U.S. program.

25. Market Loss Assistance Payments Are Non-Product-Specific Support: As indicated in the U.S. 1999 WTO domestic support notification (G/AG/N/USA/43), the expired market loss assistance payments were non-product-specific support. As with production flexibility contract payments, market loss assistance payments were made to persons with farm acres that previously had been devoted to production of certain crops, including upland cotton, during an historical base period. A recipient was not required to produce upland cotton or any other crop in order to receive payment, and no production was required at all. Thus, these payments are not product- specific support and would not form part of the Peace Clause proviso comparison.

26. Direct Payments: Were the Panel to conclude that direct payments do not conform fully to the provisions of Annex 2, direct payments would be non-product-specific support. As with counter-cyclical payments, direct payments are based on quantities of acreage that historically produced cotton, and there is no requirement to produce upland cotton (or any other crop) to receive these payments. Thus, direct payments would not be product-specific support.

27. Production Flexibility Contract Payments: Were the Panel to consider that these payments are within its terms of reference, the United States has explained that they would be green box support. Were the Panel to conclude further that production flexibility contract payments do not conform fully to the provisions of Annex 2, these payments would also be non- product-specific support for the reasons given with respect to direct payments. As such, they would not form part of the Peace Clause proviso comparison.

28. Cottonseed Payments: The Agricultural Assistance Act of 2003 and the cottonseed payment made pursuant to it is not within the Panel’s terms of reference because the legislation authorizing the payments had not even been enacted at the time of Brazil’s panel request, much less its consultation request. The “legal instruments” pursuant to which prior cottonseed payments were made, moreover, do not appear in Brazil’s consultation or panel requests. Thus, it would appear that cottonseed payments for the 1999 and 2000 crops of cottonseed also do not form part of the Panel’s terms of reference.

29. Peace Clause Comparison – The Product-Specific AMS for Upland Cotton Also Demonstrates That Challenged U.S. Measures Do Not Breach the Peace Clause: The United States believes the Peace Clause compels comparing the rate of support decided by U.S. measures, whether or not adjusted for the acreage reduction program and normal flex acres, with United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 7 the current rate of support. Were the Panel to determine to use an Aggregate Measurement of Support calculation, however, the price gap methodology is the only appropriate one for Peace Clause purposes.

30. The price gap methodology eliminates the effect of prevailing market prices on the calculation of support. Instead, paragraphs 10 and 11 of Annex 3 designate that the support be calculated by multiplying the quantity of eligible production by the gap between the applied administered price (for example, the marketing loan rate) and the fixed reference price (that is, the actual price for determining payment rates for the years 1986 to 1988). Thus, by holding the reference price “fixed,” support measured using a price gap calculation shows the effect of changes in the level of support (applied administered price) decided by a Member, rather than changes in outlays that result from movements in market prices that a Member does not control. In fact, the United States has calculated an AMS for upland cotton using the price gap methodology for both deficiency payments and marketing loan payments (marketing loan gains, certificate exchange gains, and loan deficiency payments) and using budgetary outlays for all other payments. The result is exactly the same as a rate of support comparison: in no marketing year from 1999 through 2002 is the support U.S. measures grant in excess of the 1992 marketing year level.

U.S. Export Credit Guarantee Programs

31. The Negotiating History of Article 10.2 Reveals that the Negotiators Explicitly Deferred the Application of All Export Subsidy Disciplines on Export Credit Guarantees: The GATT/WTO negotiating history regarding export credits and export credit guarantees in agriculture supports the U.S. interpretation of Article 10.2. On June 24, 1991, Chairman Dunkel circulated a Note on Options in the Agriculture Negotiations requesting decisions by the principals on “whether subsidized export credits and related practices . . . would be subject to reduction commitments.” Subsequently, on August 2, 1991, he circulated a proposed “Illustrative List of Export Subsidy Practices.” Item (h) is explicitly “Export Credits provided by governments or their agencies on less than fully commercial terms.” Similarly, item (i) is “Subsidized export credit guarantees or insurance programs.”

32. On December 20, 1991, the “Draft Final Act Embodying the Results of the of Multilateral Trade Negotiations” was issued. Article 10.2 of the Draft Final Act states: “Participants undertake not to provide export credits, export credit guarantees or insurance programs otherwise than in conformity with internationally agreed disciplines” (emphasis added). This draft text would clearly prohibit the use of export credit guarantees except in conformity with agreed disciplines. Such internationally agreed disciplines would include those contemplated by the SCM Agreement. This would be precisely the language necessary to support Brazil’s reading.

33. Ironically, Brazil’s interpretation would require export credit guarantees in agriculture to be subject to greater disciplines than any other practice addressed in the Agreement on United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 8

Agriculture. Under Brazil’s view, not only would export credit guarantees constitute export subsidies and be subject to all of the export subsidy disciplines, but Members would also be specifically obligated to work toward and then apply additional disciplines.

34. Brazil’s approach would result in gross injustice: As part of the negotiations, the parties had to prepare and submit schedules of quantities and budget outlays during a base period to derive the export subsidy reduction commitments ultimately reflected in the respective schedules of the Members. Had Members’ export credit guarantees been considered export subsidies for these purposes from the outset, then the export credit guarantee activity during the base period would also have to have been added to the base figures from which each Member’s export subsidy reduction commitments were calculated. For example, the United States has no export subsidy reduction commitment with respect to corn, yet during the 1986-1990 base period an average of over 5.5 million tons of corn were exported each year under the GSM-102 and GSM-103 programs. The United States would have reduction commitments for many more products than currently and would have had significantly increased commitments for the 13 products that are scheduled. However, Brazil would have the Panel impose the disciplines now but deny Members the corresponding changes in reduction commitments. Brazil’s approach would be grossly inequitable and the Panel should reject it.

35. The Application of Government-Wide Accounting Rules Indicates that the Export Credit Guarantee Programs are Covering Long-Term Operating Costs and Losses: The application of the Federal Credit Reform Act of 1990 (“FCRA”) over time to the export credit guarantee programs as a whole currently indicates that the net result of all activity associated with export credit guarantees issued in fiscal years 1994 and 1995 is a total net receipt to the United States of $29 million. The experience of 1994 and 1995 is viewed as representative, and the United States expects that the net results for other years will be similar to the experience for 1994 and 1995. Re-estimates thus far have resulted in a net reduction in the estimated costs of these programs of over $1.9 billion since the inception of credit reform budgeting in fiscal year 1992. Based on those results, the Brazilian claim that "operating costs and losses for GSM 102, GSM 103, and SCGP have outpaced premiums collected in every single year since the United States started applying the formula in 1992" is not supportable.

36. The United States has gathered cumulative reestimates on a cohort basis: For example, for cohort 1992 (not yet closed) the current data reflects an estimate of a profit to the United States of approximately $124 million; for 1993 (not yet closed), the corresponding current figure is a profit of approximately $56 million; and, as indicated, cohorts 1994 and 1995 together project a profit of $29 million. With the exception of 2002, for which only very recent data is necessarily available, the Panel will note that the trend for all cohorts is uniformly favorable as compared to the original subsidy estimate.

37. Brazil asserts that “historically, the majority of GSM support that is rescheduled is ‘in arrears’” and that this increases costs. Brazil largely relies, however, on a 1990 government United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 9 report that is dated and precedes FCRA itself. No rescheduling applicable to export credit guarantees issued in fiscal year 1992 or later is in arrears.

38. Brazil’s Suggestion to Use Estimated Data to Determine Long-Term Costs and Losses Supports the View that the Export Credit Guarantees Do Not Provide Export Subsidies: The United States notes Brazil’s statement that “a certain degree of estimated data would be perfectly acceptable in an analysis of the costs and losses of guarantee programs under item(j)” for two reasons. First, the re-estimate process for fiscal years 1994, 1995, and virtually every other year since fiscal year 1992 indicates a very strong net positive trend with respect to the programs and that therefore current premium rates do cover long-term operating costs and losses. Second, it is relevant with respect to Brazil’s reliance on the significant losses that the United States admittedly incurred with respect to and Iraq. Presumably, to attempt to recover such losses in any practical time frame would require such a prohibitive fee increase that few, if any, exporters would take advantage of the program. Consequently, the United States would be whipsawed by a prohibition on the export credit guarantee as currently constituted because of the large losses incurred between 10 and 20 years ago, and the inability to create a conforming program because the fee structure necessary to compensate for such historical losses would foreclose use of the program. Item (j) cannot be reasonably interpreted to require an examination of all activity since the beginning of a program, no matter how old it may be. The data provided with respect to fiscal years 1994 and 1995 and for the programs as a whole indicates that current premium rates are presently adequate to cover long-term operating costs and losses as currently projected. The United States is also in a net positive position with respect to cotton transactions in the ten years commencing with fiscal year 1993.

39. The Export Credit Guarantee Programs Are Not Applied in a Manner which Results in or which Threatens to Lead to, Circumvention of Export Subsidy Commitments: Brazil has challenged the export credit guarantee programs, GSM 102, GSM 103, and SCGP, as such. Brazil has failed, however, to demonstrate that these programs as such mandate a violation of U.S. WTO obligations. It is well established under GATT and WTO jurisprudence that legislation of a Member violates that Member’s WTO obligations only if the legislation mandates action that is inconsistent with those obligations or precludes action that is consistent with those obligations. If the legislation provides discretion to administrative authorities to act in a WTO-consistent manner, the legislation, as such, does not violate a Member’s WTO obligations. This distinction has continued under the WTO system.

40. The Commodity Credit Corporation (“CCC”) has complete statutory and regulatory discretion at any time not to issue guarantees with respect to any individual application for an export credit guarantee or to suspend the issuance of export credit guarantees under any particular allocation. This is in marked contrast to the situation in U.S.- FSC, in which the found a threat of circumvention because the FSC legislation created a legal entitlement to the payment. There is no statutory legal entitlement to an export credit guarantee. Furthermore, even if an application and fee are received, the applicant is not necessarily entitled to receive the guarantee. Issuance is discretionary. United States – Subsidies on Upland Cotton U.S. Executive Summary of Rebuttal Submission (WT/DS267) September 1, 2003 – Page 10

41. Finally, Brazil has alleged that the United States has exceeded its quantitative export subsidy reduction commitments during the period July 2001-June 2002. Even if the export credit guarantee programs were deemed export subsidies, the United States would be in compliance with the quantitative reduction commitments for that period with respect to wheat, coarse grains, butter and butter oil, skim milk powder, cheese, other milk products, bovine meat, live dairy cattle, and eggs. This may also be true with respect to vegetable oil. In fiscal year 2002, it would also be true for poultry meat. The United States did not use the GSM-102 or GSM-103 programs during 2001-2002 with respect to butter and butter oil, skim milk powder, cheese, other milk products, or eggs.

42. Financial Arrangements Analogous to the CCC Export Credit Guarantee Programs are Available in the Marketplace: In light of Article 10.2, it is neither appropriate nor necessary to analyze the export credit guarantees with respect to Article 1.1(b) of the SCM Agreement. However, we note that financing is available in the marketplace that is analogous to export credit guarantees. A prominent example in the commercial market would be “forfaiting.” It would appear, then, that a competitive marketplace exists for trade financing even in emerging markets where more conventional financing is not available. The United States is not privy to the precise terms at any time available in forfaiting transactions because those terms can vary by country, commodity, bank risk, size of transaction and numerous other factors. In addition, like most private financial activity, that information is ordinarily held confidentially by the parties.

The Step 2 Program is Not Contingent on Export Performance

43. Brazil apparently does not contest that all uses of upland cotton are eligible for the Step 2 subsidy. Instead, Brazil suggests, erroneously, that not the entire universe of users of upland cotton is eligible for the subsidy. First, the requirement that a recipient must be “regularly engaged” in the use of cotton is simply an anti-fraud provision to preclude an attempt to receive a payment with respect to cotton on which a payment has already been made. Brazil also correctly notes that “the eligible domestic user criteria exclude all firms that are domestic cotton brokers or simple resellers.” These parties are not using the cotton and are therefore ineligible. Brazil suggests a third category of persons who are users but are not eligible to receive the payment: “firms that have not entered into CCC contracts” as either manufacturers or exporters. It is true that CCC cannot pay parties that choose to remain unknown to it, but this requires an assumption of economic irrationality and does not diminish the point that all who use cotton have it entirely within their power to receive the subsidy.